1 Let us first have a discussion on the stop criterion of the citted algorithm.
2 We claim that even if the variation of the dual function is less than a given
3 threeshold, this does not ensure that the lifetime has been maximized.
4 Minimizing a function on a multiple domain (as the dual function)
5 may indeed easilly fall into a local trap because some of introduced
6 variables may lead to uniformity of the output.
10 \caption{Relations between dual function threshold and $q_i$ convergence}
11 \label{fig:convergence:scatterplot}
14 Experimentations have indeed shown that even if the dual
15 function seems to be constant
16 (variations between two evaluations of this one is less than $10^{-5}$)
17 not all the $q_i$ have the same value.
18 For instance, the Figure~\ref{fig:convergence:scatterplot} presents
21 The maximum amplitude rate of the sequence of $q_i$ --which is
22 $\frac{\max_{i \in N} q_i} {\min_{i \in N}q_i}-1$--
23 is represented in $y$-coordonates
25 value of the threeshold for dual function that is represented in
27 This figure shows that a very small threshold is a necessary condition, but not
28 a sufficient criteria to observe convergence of $q_i$.
30 In the following, we consider the system are $\epsilon$-stable if both
31 maximum amplitude rate and the dual function are less than a threeshold